Skip to content

[CI] Add file-based auto-labeling for PR components#3402

Merged
vmoens merged 4 commits intogh/vmoens/205/basefrom
gh/vmoens/205/head
Jan 29, 2026
Merged

[CI] Add file-based auto-labeling for PR components#3402
vmoens merged 4 commits intogh/vmoens/205/basefrom
gh/vmoens/205/head

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Jan 28, 2026

Stack from ghstack (oldest at bottom):

This commit adds a file-based auto-labeling system that complements the
existing title-based labeling (auto-tag.yml):

New files:

  • .github/labeler.yml: Path-to-label mappings for all TorchRL components
  • .github/workflows/auto-labeler.yml: Workflow using actions/labeler@v5

Updated files:

  • .github/labels.yml: Added missing labels for Environments (gym, sklearn,
    dm_control, habitat), TorchRL modules (Objectives, Modules, Transforms,
    Record, Trainers, WeightUpdate, Services, ReplayBuffers), and other
    categories (sota-implementations/, tutorials/)
  • .github/workflows/auto-tag.yml: Fixed to only ADD labels (never remove),
    added support for singular/plural prefix variations (e.g., [Environment]
    and [Environments] both work)

Together these provide complete PR categorization:

  • Title-based labels capture the INTENT (bug fix, feature, etc.)
  • File-based labels capture the SCOPE (which components changed)

[ghstack-poisoned]
@pytorch-bot
Copy link

pytorch-bot bot commented Jan 28, 2026

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3402

Note: Links to docs will display an error until the docs builds have been completed.

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@github-actions github-actions bot added the CI Has to do with CI setup (e.g. wheels & builds, tests...) label Jan 28, 2026
@meta-cla meta-cla bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jan 28, 2026
@github-actions
Copy link
Contributor

github-actions bot commented Jan 28, 2026

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of CPU Benchmark Tests

Total Benchmarks: 153. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}15$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 81.3864μs 80.4021μs 12.4375 KOps/s 11.7437 KOps/s $\textbf{\color{#35bf28}+5.91\%}$
test_tensor_to_bytestream_speed[torch.save] 0.1413ms 0.1406ms 7.1133 KOps/s 6.9429 KOps/s $\color{#35bf28}+2.45\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1398s 0.1398s 7.1549 Ops/s 7.2563 Ops/s $\color{#d91a1a}-1.40\%$
test_tensor_to_bytestream_speed[numpy] 2.8019μs 2.7943μs 357.8689 KOps/s 354.6395 KOps/s $\color{#35bf28}+0.91\%$
test_tensor_to_bytestream_speed[safetensors] 41.3984μs 41.1179μs 24.3203 KOps/s 27.1090 KOps/s $\textbf{\color{#d91a1a}-10.29\%}$
test_simple 0.6842s 0.5900s 1.6950 Ops/s 1.7133 Ops/s $\color{#d91a1a}-1.07\%$
test_transformed 1.2715s 1.1860s 0.8432 Ops/s 0.8515 Ops/s $\color{#d91a1a}-0.97\%$
test_serial 1.8575s 1.7707s 0.5647 Ops/s 0.5843 Ops/s $\color{#d91a1a}-3.35\%$
test_parallel 1.2608s 1.1959s 0.8362 Ops/s 0.8673 Ops/s $\color{#d91a1a}-3.59\%$
test_step_mdp_speed[True-True-True-True-True] 0.1517ms 43.8762μs 22.7914 KOps/s 22.7222 KOps/s $\color{#35bf28}+0.30\%$
test_step_mdp_speed[True-True-True-True-False] 57.7510μs 24.7936μs 40.3330 KOps/s 39.6933 KOps/s $\color{#35bf28}+1.61\%$
test_step_mdp_speed[True-True-True-False-True] 52.7310μs 24.9114μs 40.1423 KOps/s 39.0596 KOps/s $\color{#35bf28}+2.77\%$
test_step_mdp_speed[True-True-True-False-False] 38.5700μs 13.8615μs 72.1424 KOps/s 72.2167 KOps/s $\color{#d91a1a}-0.10\%$
test_step_mdp_speed[True-True-False-True-True] 75.8310μs 48.1860μs 20.7529 KOps/s 20.6710 KOps/s $\color{#35bf28}+0.40\%$
test_step_mdp_speed[True-True-False-True-False] 53.8010μs 27.8085μs 35.9602 KOps/s 35.5609 KOps/s $\color{#35bf28}+1.12\%$
test_step_mdp_speed[True-True-False-False-True] 56.9210μs 28.1740μs 35.4937 KOps/s 35.2817 KOps/s $\color{#35bf28}+0.60\%$
test_step_mdp_speed[True-True-False-False-False] 85.9120μs 16.4744μs 60.7001 KOps/s 59.9629 KOps/s $\color{#35bf28}+1.23\%$
test_step_mdp_speed[True-False-True-True-True] 88.0820μs 51.2599μs 19.5084 KOps/s 19.9133 KOps/s $\color{#d91a1a}-2.03\%$
test_step_mdp_speed[True-False-True-True-False] 78.5820μs 30.7406μs 32.5303 KOps/s 32.1874 KOps/s $\color{#35bf28}+1.07\%$
test_step_mdp_speed[True-False-True-False-True] 91.0420μs 27.7730μs 36.0062 KOps/s 35.1463 KOps/s $\color{#35bf28}+2.45\%$
test_step_mdp_speed[True-False-True-False-False] 47.6310μs 16.6194μs 60.1705 KOps/s 61.0348 KOps/s $\color{#d91a1a}-1.42\%$
test_step_mdp_speed[True-False-False-True-True] 87.0820μs 53.3257μs 18.7527 KOps/s 18.8400 KOps/s $\color{#d91a1a}-0.46\%$
test_step_mdp_speed[True-False-False-True-False] 75.5220μs 32.9410μs 30.3573 KOps/s 30.3257 KOps/s $\color{#35bf28}+0.10\%$
test_step_mdp_speed[True-False-False-False-True] 61.9810μs 30.8209μs 32.4455 KOps/s 32.9592 KOps/s $\color{#d91a1a}-1.56\%$
test_step_mdp_speed[True-False-False-False-False] 52.8220μs 19.3291μs 51.7354 KOps/s 52.9247 KOps/s $\color{#d91a1a}-2.25\%$
test_step_mdp_speed[False-True-True-True-True] 92.8920μs 49.4764μs 20.2117 KOps/s 19.9077 KOps/s $\color{#35bf28}+1.53\%$
test_step_mdp_speed[False-True-True-True-False] 61.5810μs 29.8968μs 33.4484 KOps/s 33.1147 KOps/s $\color{#35bf28}+1.01\%$
test_step_mdp_speed[False-True-True-False-True] 0.1002ms 31.7096μs 31.5362 KOps/s 31.2789 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[False-True-True-False-False] 39.7010μs 18.1426μs 55.1189 KOps/s 54.7791 KOps/s $\color{#35bf28}+0.62\%$
test_step_mdp_speed[False-True-False-True-True] 2.7141ms 52.8439μs 18.9236 KOps/s 18.8912 KOps/s $\color{#35bf28}+0.17\%$
test_step_mdp_speed[False-True-False-True-False] 78.8020μs 33.0602μs 30.2479 KOps/s 30.3460 KOps/s $\color{#d91a1a}-0.32\%$
test_step_mdp_speed[False-True-False-False-True] 60.1520μs 34.4585μs 29.0204 KOps/s 28.8992 KOps/s $\color{#35bf28}+0.42\%$
test_step_mdp_speed[False-True-False-False-False] 55.4810μs 20.8666μs 47.9234 KOps/s 48.4124 KOps/s $\color{#d91a1a}-1.01\%$
test_step_mdp_speed[False-False-True-True-True] 0.1002ms 55.2421μs 18.1021 KOps/s 17.7797 KOps/s $\color{#35bf28}+1.81\%$
test_step_mdp_speed[False-False-True-True-False] 64.3310μs 36.1352μs 27.6738 KOps/s 27.5056 KOps/s $\color{#35bf28}+0.61\%$
test_step_mdp_speed[False-False-True-False-True] 67.8120μs 33.9345μs 29.4686 KOps/s 29.3391 KOps/s $\color{#35bf28}+0.44\%$
test_step_mdp_speed[False-False-True-False-False] 53.6410μs 20.9521μs 47.7278 KOps/s 48.3184 KOps/s $\color{#d91a1a}-1.22\%$
test_step_mdp_speed[False-False-False-True-True] 0.1061ms 57.9112μs 17.2678 KOps/s 17.1820 KOps/s $\color{#35bf28}+0.50\%$
test_step_mdp_speed[False-False-False-True-False] 73.4710μs 38.5468μs 25.9425 KOps/s 26.2174 KOps/s $\color{#d91a1a}-1.05\%$
test_step_mdp_speed[False-False-False-False-True] 71.7310μs 35.6429μs 28.0561 KOps/s 27.3630 KOps/s $\color{#35bf28}+2.53\%$
test_step_mdp_speed[False-False-False-False-False] 52.6610μs 23.4673μs 42.6126 KOps/s 42.9632 KOps/s $\color{#d91a1a}-0.82\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8710s 0.7731s 1.2936 Ops/s 1.2675 Ops/s $\color{#35bf28}+2.06\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7340s 0.6395s 1.5636 Ops/s 1.5472 Ops/s $\color{#35bf28}+1.06\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7626s 1.6882s 0.5924 Ops/s 0.5882 Ops/s $\color{#35bf28}+0.71\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5349s 1.4624s 0.6838 Ops/s 0.6783 Ops/s $\color{#35bf28}+0.81\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0196s 1.9440s 0.5144 Ops/s 0.5137 Ops/s $\color{#35bf28}+0.14\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7920s 1.7177s 0.5822 Ops/s 0.5801 Ops/s $\color{#35bf28}+0.35\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.8039s 4.6742s 0.2139 Ops/s 0.2112 Ops/s $\color{#35bf28}+1.28\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.6330s 4.5346s 0.2205 Ops/s 0.2206 Ops/s $\color{#d91a1a}-0.05\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.1966s 2.0460s 0.4887 Ops/s 0.5007 Ops/s $\color{#d91a1a}-2.38\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7645s 1.6971s 0.5892 Ops/s 0.5690 Ops/s $\color{#35bf28}+3.56\%$
test_values[generalized_advantage_estimate-True-True] 11.4564ms 11.2755ms 88.6875 Ops/s 87.3381 Ops/s $\color{#35bf28}+1.55\%$
test_values[vec_generalized_advantage_estimate-True-True] 19.7183ms 17.9139ms 55.8227 Ops/s 88.2222 Ops/s $\textbf{\color{#d91a1a}-36.72\%}$
test_values[td0_return_estimate-False-False] 0.2253ms 0.1315ms 7.6065 KOps/s 7.4658 KOps/s $\color{#35bf28}+1.88\%$
test_values[td1_return_estimate-False-False] 31.2864ms 30.7479ms 32.5225 Ops/s 32.1841 Ops/s $\color{#35bf28}+1.05\%$
test_values[vec_td1_return_estimate-False-False] 18.3544ms 17.9260ms 55.7850 Ops/s 87.5037 Ops/s $\textbf{\color{#d91a1a}-36.25\%}$
test_values[td_lambda_return_estimate-True-False] 45.7595ms 45.2766ms 22.0864 Ops/s 21.6524 Ops/s $\color{#35bf28}+2.00\%$
test_values[vec_td_lambda_return_estimate-True-False] 18.4191ms 17.8888ms 55.9008 Ops/s 87.6872 Ops/s $\textbf{\color{#d91a1a}-36.25\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 10.3690ms 10.1943ms 98.0942 Ops/s 96.8479 Ops/s $\color{#35bf28}+1.29\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7989ms 1.5805ms 632.7163 Ops/s 648.9942 Ops/s $\color{#d91a1a}-2.51\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4926ms 0.4398ms 2.2737 KOps/s 2.2768 KOps/s $\color{#d91a1a}-0.14\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 35.2780ms 34.7877ms 28.7458 Ops/s 34.1161 Ops/s $\textbf{\color{#d91a1a}-15.74\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.8927ms 1.7713ms 564.5483 Ops/s 560.2168 Ops/s $\color{#35bf28}+0.77\%$
test_dqn_speed[False-None] 2.0015ms 1.4665ms 681.8916 Ops/s 693.8282 Ops/s $\color{#d91a1a}-1.72\%$
test_dqn_speed[False-backward] 2.0199ms 1.9653ms 508.8153 Ops/s 454.1623 Ops/s $\textbf{\color{#35bf28}+12.03\%}$
test_dqn_speed[True-None] 0.7081ms 0.5440ms 1.8381 KOps/s 1.6802 KOps/s $\textbf{\color{#35bf28}+9.40\%}$
test_dqn_speed[True-backward] 1.0622ms 1.0122ms 987.9153 Ops/s 892.4390 Ops/s $\textbf{\color{#35bf28}+10.70\%}$
test_dqn_speed[reduce-overhead-None] 0.5866ms 0.5359ms 1.8661 KOps/s 1.7877 KOps/s $\color{#35bf28}+4.39\%$
test_ddpg_speed[False-None] 0.2005s 3.5158ms 284.4317 Ops/s 336.7463 Ops/s $\textbf{\color{#d91a1a}-15.54\%}$
test_ddpg_speed[False-backward] 4.2261ms 4.1363ms 241.7607 Ops/s 239.3635 Ops/s $\color{#35bf28}+1.00\%$
test_ddpg_speed[True-None] 1.7955ms 1.4224ms 703.0237 Ops/s 689.1671 Ops/s $\color{#35bf28}+2.01\%$
test_ddpg_speed[True-backward] 2.8349ms 2.4482ms 408.4694 Ops/s 370.4421 Ops/s $\textbf{\color{#35bf28}+10.27\%}$
test_ddpg_speed[reduce-overhead-None] 1.5346ms 1.4230ms 702.7426 Ops/s 697.2706 Ops/s $\color{#35bf28}+0.78\%$
test_sac_speed[False-None] 8.7628ms 8.2252ms 121.5769 Ops/s 121.2002 Ops/s $\color{#35bf28}+0.31\%$
test_sac_speed[False-backward] 12.1971ms 11.5010ms 86.9493 Ops/s 86.4219 Ops/s $\color{#35bf28}+0.61\%$
test_sac_speed[True-None] 2.5675ms 2.1755ms 459.6559 Ops/s 431.3400 Ops/s $\textbf{\color{#35bf28}+6.56\%}$
test_sac_speed[True-backward] 4.1210ms 4.0505ms 246.8834 Ops/s 220.3671 Ops/s $\textbf{\color{#35bf28}+12.03\%}$
test_sac_speed[reduce-overhead-None] 2.2347ms 2.1400ms 467.2937 Ops/s 437.8173 Ops/s $\textbf{\color{#35bf28}+6.73\%}$
test_redq_speed[False-None] 17.0604ms 11.0506ms 90.4932 Ops/s 96.0761 Ops/s $\textbf{\color{#d91a1a}-5.81\%}$
test_redq_speed[False-backward] 18.6626ms 18.0088ms 55.5284 Ops/s 56.2307 Ops/s $\color{#d91a1a}-1.25\%$
test_redq_speed[True-None] 4.7493ms 4.4702ms 223.7014 Ops/s 222.5831 Ops/s $\color{#35bf28}+0.50\%$
test_redq_speed[True-backward] 10.0269ms 9.8618ms 101.4012 Ops/s 99.2247 Ops/s $\color{#35bf28}+2.19\%$
test_redq_speed[reduce-overhead-None] 4.9524ms 4.3968ms 227.4382 Ops/s 218.8329 Ops/s $\color{#35bf28}+3.93\%$
test_redq_deprec_speed[False-None] 11.5279ms 11.2108ms 89.2000 Ops/s 89.8085 Ops/s $\color{#d91a1a}-0.68\%$
test_redq_deprec_speed[False-backward] 16.4261ms 16.1188ms 62.0392 Ops/s 62.9828 Ops/s $\color{#d91a1a}-1.50\%$
test_redq_deprec_speed[True-None] 3.8562ms 3.7047ms 269.9242 Ops/s 277.5423 Ops/s $\color{#d91a1a}-2.74\%$
test_redq_deprec_speed[True-backward] 7.9893ms 7.7088ms 129.7213 Ops/s 121.1946 Ops/s $\textbf{\color{#35bf28}+7.04\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.8709ms 3.6302ms 275.4674 Ops/s 263.0786 Ops/s $\color{#35bf28}+4.71\%$
test_td3_speed[False-None] 8.2839ms 8.1709ms 122.3857 Ops/s 121.6576 Ops/s $\color{#35bf28}+0.60\%$
test_td3_speed[False-backward] 11.5122ms 11.0731ms 90.3091 Ops/s 89.6599 Ops/s $\color{#35bf28}+0.72\%$
test_td3_speed[True-None] 1.9168ms 1.8562ms 538.7226 Ops/s 528.7064 Ops/s $\color{#35bf28}+1.89\%$
test_td3_speed[True-backward] 3.7663ms 3.6707ms 272.4287 Ops/s 220.3274 Ops/s $\textbf{\color{#35bf28}+23.65\%}$
test_td3_speed[reduce-overhead-None] 1.8565ms 1.8090ms 552.7900 Ops/s 536.2042 Ops/s $\color{#35bf28}+3.09\%$
test_cql_speed[False-None] 29.6680ms 26.5371ms 37.6830 Ops/s 37.8246 Ops/s $\color{#d91a1a}-0.37\%$
test_cql_speed[False-backward] 36.3426ms 35.6252ms 28.0700 Ops/s 28.0274 Ops/s $\color{#35bf28}+0.15\%$
test_cql_speed[True-None] 12.7971ms 12.6222ms 79.2254 Ops/s 77.6115 Ops/s $\color{#35bf28}+2.08\%$
test_cql_speed[True-backward] 19.2187ms 18.6047ms 53.7498 Ops/s 53.2098 Ops/s $\color{#35bf28}+1.01\%$
test_cql_speed[reduce-overhead-None] 12.7410ms 12.5190ms 79.8789 Ops/s 78.5422 Ops/s $\color{#35bf28}+1.70\%$
test_a2c_speed[False-None] 5.7819ms 5.5664ms 179.6493 Ops/s 182.5028 Ops/s $\color{#d91a1a}-1.56\%$
test_a2c_speed[False-backward] 12.2615ms 11.9976ms 83.3501 Ops/s 83.8437 Ops/s $\color{#d91a1a}-0.59\%$
test_a2c_speed[True-None] 3.9327ms 3.7348ms 267.7510 Ops/s 258.6815 Ops/s $\color{#35bf28}+3.51\%$
test_a2c_speed[True-backward] 8.8038ms 8.6236ms 115.9602 Ops/s 113.3873 Ops/s $\color{#35bf28}+2.27\%$
test_a2c_speed[reduce-overhead-None] 3.8256ms 3.7100ms 269.5411 Ops/s 267.1865 Ops/s $\color{#35bf28}+0.88\%$
test_ppo_speed[False-None] 6.1680ms 5.9924ms 166.8781 Ops/s 163.7666 Ops/s $\color{#35bf28}+1.90\%$
test_ppo_speed[False-backward] 12.9694ms 12.7177ms 78.6305 Ops/s 79.2821 Ops/s $\color{#d91a1a}-0.82\%$
test_ppo_speed[True-None] 3.7566ms 3.6281ms 275.6251 Ops/s 272.5474 Ops/s $\color{#35bf28}+1.13\%$
test_ppo_speed[True-backward] 8.7255ms 8.4277ms 118.6562 Ops/s 113.7371 Ops/s $\color{#35bf28}+4.32\%$
test_ppo_speed[reduce-overhead-None] 3.7597ms 3.6062ms 277.3039 Ops/s 276.6578 Ops/s $\color{#35bf28}+0.23\%$
test_reinforce_speed[False-None] 5.4237ms 4.6305ms 215.9576 Ops/s 219.8256 Ops/s $\color{#d91a1a}-1.76\%$
test_reinforce_speed[False-backward] 8.5498ms 7.4615ms 134.0215 Ops/s 136.1514 Ops/s $\color{#d91a1a}-1.56\%$
test_reinforce_speed[True-None] 3.1507ms 2.8903ms 345.9905 Ops/s 345.7848 Ops/s $\color{#35bf28}+0.06\%$
test_reinforce_speed[True-backward] 8.1168ms 7.8812ms 126.8838 Ops/s 128.4643 Ops/s $\color{#d91a1a}-1.23\%$
test_reinforce_speed[reduce-overhead-None] 3.0907ms 2.8900ms 346.0250 Ops/s 343.9929 Ops/s $\color{#35bf28}+0.59\%$
test_iql_speed[False-None] 27.0593ms 20.7679ms 48.1513 Ops/s 49.7646 Ops/s $\color{#d91a1a}-3.24\%$
test_iql_speed[False-backward] 36.4681ms 30.9354ms 32.3254 Ops/s 32.6118 Ops/s $\color{#d91a1a}-0.88\%$
test_iql_speed[True-None] 8.7565ms 8.5301ms 117.2318 Ops/s 110.7362 Ops/s $\textbf{\color{#35bf28}+5.87\%}$
test_iql_speed[True-backward] 17.0999ms 16.7688ms 59.6344 Ops/s 58.3241 Ops/s $\color{#35bf28}+2.25\%$
test_iql_speed[reduce-overhead-None] 8.8786ms 8.5990ms 116.2927 Ops/s 113.3755 Ops/s $\color{#35bf28}+2.57\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.3195ms 6.1585ms 162.3774 Ops/s 165.9231 Ops/s $\color{#d91a1a}-2.14\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 3.2304ms 0.3431ms 2.9144 KOps/s 2.8090 KOps/s $\color{#35bf28}+3.75\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7056ms 0.3702ms 2.7014 KOps/s 2.8951 KOps/s $\textbf{\color{#d91a1a}-6.69\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2123ms 5.9890ms 166.9740 Ops/s 171.1060 Ops/s $\color{#d91a1a}-2.41\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9993ms 0.3737ms 2.6759 KOps/s 2.8857 KOps/s $\textbf{\color{#d91a1a}-7.27\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5714ms 0.3614ms 2.7672 KOps/s 3.0266 KOps/s $\textbf{\color{#d91a1a}-8.57\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6920ms 1.4800ms 675.6776 Ops/s 680.8009 Ops/s $\color{#d91a1a}-0.75\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.6059ms 1.3952ms 716.7539 Ops/s 719.8911 Ops/s $\color{#d91a1a}-0.44\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2005ms 6.0550ms 165.1519 Ops/s 166.5800 Ops/s $\color{#d91a1a}-0.86\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8724ms 0.4698ms 2.1285 KOps/s 2.0746 KOps/s $\color{#35bf28}+2.60\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6809ms 0.4782ms 2.0912 KOps/s 2.0184 KOps/s $\color{#35bf28}+3.60\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1234ms 5.9635ms 167.6876 Ops/s 169.6555 Ops/s $\color{#d91a1a}-1.16\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9546ms 0.3396ms 2.9447 KOps/s 2.9003 KOps/s $\color{#35bf28}+1.53\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6578ms 0.3642ms 2.7456 KOps/s 3.1136 KOps/s $\textbf{\color{#d91a1a}-11.82\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2736ms 5.9409ms 168.3251 Ops/s 171.6562 Ops/s $\color{#d91a1a}-1.94\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8207ms 0.3637ms 2.7493 KOps/s 3.2162 KOps/s $\textbf{\color{#d91a1a}-14.52\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5393ms 0.3294ms 3.0358 KOps/s 3.4018 KOps/s $\textbf{\color{#d91a1a}-10.76\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.1733ms 6.0909ms 164.1789 Ops/s 166.3585 Ops/s $\color{#d91a1a}-1.31\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.5424ms 0.5232ms 1.9115 KOps/s 1.8677 KOps/s $\color{#35bf28}+2.35\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9229ms 0.4415ms 2.2648 KOps/s 1.9244 KOps/s $\textbf{\color{#35bf28}+17.69\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.5909s 16.8594ms 59.3140 Ops/s 193.7160 Ops/s $\textbf{\color{#d91a1a}-69.38\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 9.3260ms 1.9836ms 504.1353 Ops/s 497.3592 Ops/s $\color{#35bf28}+1.36\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.1982ms 0.8772ms 1.1400 KOps/s 1.1135 KOps/s $\color{#35bf28}+2.38\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 8.0403ms 5.1725ms 193.3300 Ops/s 196.2312 Ops/s $\color{#d91a1a}-1.48\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 10.7370ms 1.9410ms 515.2082 Ops/s 528.3595 Ops/s $\color{#d91a1a}-2.49\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 8.3693ms 1.2723ms 786.0000 Ops/s 1.1294 KOps/s $\textbf{\color{#d91a1a}-30.41\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5516s 16.2207ms 61.6497 Ops/s 56.5843 Ops/s $\textbf{\color{#35bf28}+8.95\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 12.8898ms 2.1346ms 468.4620 Ops/s 483.6357 Ops/s $\color{#d91a1a}-3.14\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 1.4184ms 1.0618ms 941.8316 Ops/s 953.5614 Ops/s $\color{#d91a1a}-1.23\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 38.6333ms 36.4672ms 27.4219 Ops/s 27.1045 Ops/s $\color{#35bf28}+1.17\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 20.0830ms 18.6468ms 53.6286 Ops/s 53.6380 Ops/s $\color{#d91a1a}-0.02\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 41.4793ms 37.6907ms 26.5317 Ops/s 26.3199 Ops/s $\color{#35bf28}+0.80\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 20.7475ms 18.9652ms 52.7282 Ops/s 52.7643 Ops/s $\color{#d91a1a}-0.07\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 41.0643ms 39.5863ms 25.2613 Ops/s 25.1200 Ops/s $\color{#35bf28}+0.56\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 23.9289ms 20.7042ms 48.2994 Ops/s 48.5377 Ops/s $\color{#d91a1a}-0.49\%$

@github-actions
Copy link
Contributor

github-actions bot commented Jan 28, 2026

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of GPU Benchmark Tests

Total Benchmarks: 148. Improved: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 80.2894μs 79.2353μs 12.6206 KOps/s 12.4850 KOps/s $\color{#35bf28}+1.09\%$
test_tensor_to_bytestream_speed[torch.save] 0.1376ms 0.1367ms 7.3146 KOps/s 7.2105 KOps/s $\color{#35bf28}+1.44\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1116s 0.1112s 8.9894 Ops/s 8.9739 Ops/s $\color{#35bf28}+0.17\%$
test_tensor_to_bytestream_speed[numpy] 2.6364μs 2.6350μs 379.5126 KOps/s 393.3282 KOps/s $\color{#d91a1a}-3.51\%$
test_tensor_to_bytestream_speed[safetensors] 38.1984μs 36.8115μs 27.1654 KOps/s 28.2732 KOps/s $\color{#d91a1a}-3.92\%$
test_simple 0.8960s 0.8062s 1.2404 Ops/s 1.2751 Ops/s $\color{#d91a1a}-2.72\%$
test_transformed 1.5140s 1.4229s 0.7028 Ops/s 0.7111 Ops/s $\color{#d91a1a}-1.16\%$
test_serial 2.3765s 2.2841s 0.4378 Ops/s 0.4420 Ops/s $\color{#d91a1a}-0.95\%$
test_parallel 2.1790s 2.0233s 0.4942 Ops/s 0.5275 Ops/s $\textbf{\color{#d91a1a}-6.31\%}$
test_step_mdp_speed[True-True-True-True-True] 0.2797ms 41.9044μs 23.8638 KOps/s 22.5840 KOps/s $\textbf{\color{#35bf28}+5.67\%}$
test_step_mdp_speed[True-True-True-True-False] 81.2420μs 23.8462μs 41.9354 KOps/s 40.8064 KOps/s $\color{#35bf28}+2.77\%$
test_step_mdp_speed[True-True-True-False-True] 52.3610μs 23.4365μs 42.6685 KOps/s 40.8463 KOps/s $\color{#35bf28}+4.46\%$
test_step_mdp_speed[True-True-True-False-False] 39.2400μs 13.0676μs 76.5251 KOps/s 74.5657 KOps/s $\color{#35bf28}+2.63\%$
test_step_mdp_speed[True-True-False-True-True] 88.8120μs 45.0207μs 22.2120 KOps/s 21.4342 KOps/s $\color{#35bf28}+3.63\%$
test_step_mdp_speed[True-True-False-True-False] 60.5320μs 26.4054μs 37.8711 KOps/s 36.6309 KOps/s $\color{#35bf28}+3.39\%$
test_step_mdp_speed[True-True-False-False-True] 62.1710μs 26.2632μs 38.0761 KOps/s 37.2578 KOps/s $\color{#35bf28}+2.20\%$
test_step_mdp_speed[True-True-False-False-False] 59.0210μs 15.6437μs 63.9234 KOps/s 61.4546 KOps/s $\color{#35bf28}+4.02\%$
test_step_mdp_speed[True-False-True-True-True] 0.1182ms 47.5051μs 21.0504 KOps/s 20.4941 KOps/s $\color{#35bf28}+2.71\%$
test_step_mdp_speed[True-False-True-True-False] 62.9210μs 29.0146μs 34.4654 KOps/s 33.8294 KOps/s $\color{#35bf28}+1.88\%$
test_step_mdp_speed[True-False-True-False-True] 84.1310μs 26.2536μs 38.0900 KOps/s 37.7405 KOps/s $\color{#35bf28}+0.93\%$
test_step_mdp_speed[True-False-True-False-False] 64.4220μs 15.6404μs 63.9368 KOps/s 61.7718 KOps/s $\color{#35bf28}+3.50\%$
test_step_mdp_speed[True-False-False-True-True] 78.3210μs 49.8069μs 20.0775 KOps/s 19.4567 KOps/s $\color{#35bf28}+3.19\%$
test_step_mdp_speed[True-False-False-True-False] 68.2020μs 31.3078μs 31.9409 KOps/s 31.0953 KOps/s $\color{#35bf28}+2.72\%$
test_step_mdp_speed[True-False-False-False-True] 67.1410μs 28.3596μs 35.2614 KOps/s 33.9704 KOps/s $\color{#35bf28}+3.80\%$
test_step_mdp_speed[True-False-False-False-False] 44.1810μs 18.1606μs 55.0643 KOps/s 52.9911 KOps/s $\color{#35bf28}+3.91\%$
test_step_mdp_speed[False-True-True-True-True] 0.1016ms 47.7845μs 20.9273 KOps/s 20.3527 KOps/s $\color{#35bf28}+2.82\%$
test_step_mdp_speed[False-True-True-True-False] 63.8110μs 29.1776μs 34.2729 KOps/s 33.4713 KOps/s $\color{#35bf28}+2.39\%$
test_step_mdp_speed[False-True-True-False-True] 68.3010μs 29.9629μs 33.3747 KOps/s 32.9955 KOps/s $\color{#35bf28}+1.15\%$
test_step_mdp_speed[False-True-True-False-False] 47.1800μs 17.3613μs 57.5993 KOps/s 56.4045 KOps/s $\color{#35bf28}+2.12\%$
test_step_mdp_speed[False-True-False-True-True] 2.7333ms 49.9544μs 20.0183 KOps/s 19.1554 KOps/s $\color{#35bf28}+4.50\%$
test_step_mdp_speed[False-True-False-True-False] 76.2310μs 31.7286μs 31.5173 KOps/s 30.5161 KOps/s $\color{#35bf28}+3.28\%$
test_step_mdp_speed[False-True-False-False-True] 77.7310μs 32.2211μs 31.0356 KOps/s 30.2167 KOps/s $\color{#35bf28}+2.71\%$
test_step_mdp_speed[False-True-False-False-False] 63.5910μs 19.9983μs 50.0043 KOps/s 48.3531 KOps/s $\color{#35bf28}+3.41\%$
test_step_mdp_speed[False-False-True-True-True] 90.1820μs 53.0889μs 18.8363 KOps/s 18.6804 KOps/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[False-False-True-True-False] 93.3520μs 34.3688μs 29.0962 KOps/s 27.9575 KOps/s $\color{#35bf28}+4.07\%$
test_step_mdp_speed[False-False-True-False-True] 87.9810μs 32.2645μs 30.9938 KOps/s 30.2387 KOps/s $\color{#35bf28}+2.50\%$
test_step_mdp_speed[False-False-True-False-False] 52.9910μs 19.9822μs 50.0446 KOps/s 49.2547 KOps/s $\color{#35bf28}+1.60\%$
test_step_mdp_speed[False-False-False-True-True] 92.6020μs 54.6905μs 18.2847 KOps/s 17.7343 KOps/s $\color{#35bf28}+3.10\%$
test_step_mdp_speed[False-False-False-True-False] 87.7520μs 36.5335μs 27.3721 KOps/s 26.2756 KOps/s $\color{#35bf28}+4.17\%$
test_step_mdp_speed[False-False-False-False-True] 70.9510μs 34.3405μs 29.1202 KOps/s 28.5665 KOps/s $\color{#35bf28}+1.94\%$
test_step_mdp_speed[False-False-False-False-False] 51.6410μs 22.3432μs 44.7563 KOps/s 44.6010 KOps/s $\color{#35bf28}+0.35\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8339s 0.7366s 1.3577 Ops/s 1.3337 Ops/s $\color{#35bf28}+1.80\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.6989s 0.6029s 1.6587 Ops/s 1.6253 Ops/s $\color{#35bf28}+2.06\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.6655s 1.5925s 0.6279 Ops/s 0.6093 Ops/s $\color{#35bf28}+3.06\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.4555s 1.3792s 0.7251 Ops/s 0.7026 Ops/s $\color{#35bf28}+3.20\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 1.9084s 1.8283s 0.5470 Ops/s 0.5291 Ops/s $\color{#35bf28}+3.37\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.6952s 1.6176s 0.6182 Ops/s 0.5994 Ops/s $\color{#35bf28}+3.13\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.6871s 4.5560s 0.2195 Ops/s 0.2194 Ops/s $\color{#35bf28}+0.02\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.4542s 4.3539s 0.2297 Ops/s 0.2271 Ops/s $\color{#35bf28}+1.14\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 1.9762s 1.8923s 0.5285 Ops/s 0.5295 Ops/s $\color{#d91a1a}-0.20\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.6976s 1.6120s 0.6204 Ops/s 0.6084 Ops/s $\color{#35bf28}+1.97\%$
test_values[generalized_advantage_estimate-True-True] 20.2811ms 19.8814ms 50.2983 Ops/s 48.4174 Ops/s $\color{#35bf28}+3.88\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1306s 3.5281ms 283.4418 Ops/s 259.5360 Ops/s $\textbf{\color{#35bf28}+9.21\%}$
test_values[td0_return_estimate-False-False] 0.1050ms 82.2947μs 12.1515 KOps/s 12.0190 KOps/s $\color{#35bf28}+1.10\%$
test_values[td1_return_estimate-False-False] 47.9008ms 47.5709ms 21.0213 Ops/s 20.1001 Ops/s $\color{#35bf28}+4.58\%$
test_values[vec_td1_return_estimate-False-False] 1.2941ms 1.0845ms 922.0516 Ops/s 915.7262 Ops/s $\color{#35bf28}+0.69\%$
test_values[td_lambda_return_estimate-True-False] 82.0295ms 78.1842ms 12.7903 Ops/s 12.4726 Ops/s $\color{#35bf28}+2.55\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2404ms 1.0807ms 925.3267 Ops/s 919.1560 Ops/s $\color{#35bf28}+0.67\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 21.7076ms 20.9034ms 47.8391 Ops/s 48.5417 Ops/s $\color{#d91a1a}-1.45\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0593ms 0.7555ms 1.3236 KOps/s 1.3130 KOps/s $\color{#35bf28}+0.81\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7333ms 0.6783ms 1.4743 KOps/s 1.4705 KOps/s $\color{#35bf28}+0.26\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5687ms 1.4981ms 667.5126 Ops/s 669.9137 Ops/s $\color{#d91a1a}-0.36\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7608ms 0.6941ms 1.4407 KOps/s 1.4103 KOps/s $\color{#35bf28}+2.15\%$
test_dqn_speed[False-None] 1.6101ms 1.5185ms 658.5662 Ops/s 660.3988 Ops/s $\color{#d91a1a}-0.28\%$
test_dqn_speed[False-backward] 2.4547ms 2.1567ms 463.6646 Ops/s 457.0051 Ops/s $\color{#35bf28}+1.46\%$
test_dqn_speed[True-None] 0.6009ms 0.5306ms 1.8848 KOps/s 1.8515 KOps/s $\color{#35bf28}+1.80\%$
test_dqn_speed[True-backward] 1.2075ms 1.1606ms 861.6051 Ops/s 852.5750 Ops/s $\color{#35bf28}+1.06\%$
test_dqn_speed[reduce-overhead-None] 0.6378ms 0.5757ms 1.7371 KOps/s 1.7244 KOps/s $\color{#35bf28}+0.73\%$
test_ddpg_speed[False-None] 3.2039ms 2.8994ms 344.8999 Ops/s 349.1360 Ops/s $\color{#d91a1a}-1.21\%$
test_ddpg_speed[False-backward] 4.7269ms 4.2944ms 232.8588 Ops/s 233.3197 Ops/s $\color{#d91a1a}-0.20\%$
test_ddpg_speed[True-None] 1.4118ms 1.2740ms 784.9224 Ops/s 783.0212 Ops/s $\color{#35bf28}+0.24\%$
test_ddpg_speed[True-backward] 2.5652ms 2.4428ms 409.3711 Ops/s 405.7271 Ops/s $\color{#35bf28}+0.90\%$
test_ddpg_speed[reduce-overhead-None] 1.3905ms 1.3060ms 765.7145 Ops/s 762.1383 Ops/s $\color{#35bf28}+0.47\%$
test_sac_speed[False-None] 8.6955ms 8.2411ms 121.3424 Ops/s 120.1062 Ops/s $\color{#35bf28}+1.03\%$
test_sac_speed[False-backward] 11.8621ms 11.4619ms 87.2454 Ops/s 86.1663 Ops/s $\color{#35bf28}+1.25\%$
test_sac_speed[True-None] 1.8413ms 1.7548ms 569.8759 Ops/s 555.4874 Ops/s $\color{#35bf28}+2.59\%$
test_sac_speed[True-backward] 3.5267ms 3.4699ms 288.1944 Ops/s 301.2315 Ops/s $\color{#d91a1a}-4.33\%$
test_sac_speed[reduce-overhead-None] 18.5050ms 10.5962ms 94.3738 Ops/s 96.6347 Ops/s $\color{#d91a1a}-2.34\%$
test_redq_deprec_speed[False-None] 10.0494ms 9.2435ms 108.1839 Ops/s 106.3315 Ops/s $\color{#35bf28}+1.74\%$
test_redq_deprec_speed[False-backward] 13.0049ms 12.5749ms 79.5235 Ops/s 79.4720 Ops/s $\color{#35bf28}+0.06\%$
test_redq_deprec_speed[True-None] 2.5654ms 2.4772ms 403.6887 Ops/s 398.1329 Ops/s $\color{#35bf28}+1.40\%$
test_redq_deprec_speed[True-backward] 4.6128ms 4.1932ms 238.4797 Ops/s 246.6890 Ops/s $\color{#d91a1a}-3.33\%$
test_redq_deprec_speed[reduce-overhead-None] 15.4608ms 9.4963ms 105.3042 Ops/s 105.7557 Ops/s $\color{#d91a1a}-0.43\%$
test_td3_speed[False-None] 8.2774ms 8.1468ms 122.7475 Ops/s 121.8888 Ops/s $\color{#35bf28}+0.70\%$
test_td3_speed[False-backward] 11.1847ms 10.7784ms 92.7777 Ops/s 94.1251 Ops/s $\color{#d91a1a}-1.43\%$
test_td3_speed[True-None] 1.6152ms 1.5895ms 629.1341 Ops/s 632.3975 Ops/s $\color{#d91a1a}-0.52\%$
test_td3_speed[True-backward] 3.1799ms 3.1382ms 318.6516 Ops/s 328.6403 Ops/s $\color{#d91a1a}-3.04\%$
test_td3_speed[reduce-overhead-None] 55.4300ms 23.1018ms 43.2866 Ops/s 43.8876 Ops/s $\color{#d91a1a}-1.37\%$
test_cql_speed[False-None] 18.0745ms 17.2039ms 58.1263 Ops/s 58.1880 Ops/s $\color{#d91a1a}-0.11\%$
test_cql_speed[False-backward] 23.4609ms 22.7672ms 43.9228 Ops/s 44.4902 Ops/s $\color{#d91a1a}-1.28\%$
test_cql_speed[True-None] 3.2211ms 3.1695ms 315.5088 Ops/s 315.6985 Ops/s $\color{#d91a1a}-0.06\%$
test_cql_speed[True-backward] 5.7666ms 5.3793ms 185.8966 Ops/s 191.4964 Ops/s $\color{#d91a1a}-2.92\%$
test_cql_speed[reduce-overhead-None] 18.8768ms 11.6360ms 85.9405 Ops/s 86.3608 Ops/s $\color{#d91a1a}-0.49\%$
test_a2c_speed[False-None] 4.4023ms 3.2223ms 310.3361 Ops/s 308.2361 Ops/s $\color{#35bf28}+0.68\%$
test_a2c_speed[False-backward] 6.7725ms 6.3600ms 157.2325 Ops/s 161.7622 Ops/s $\color{#d91a1a}-2.80\%$
test_a2c_speed[True-None] 1.3579ms 1.3037ms 767.0632 Ops/s 775.4529 Ops/s $\color{#d91a1a}-1.08\%$
test_a2c_speed[True-backward] 3.0641ms 3.0128ms 331.9127 Ops/s 331.3054 Ops/s $\color{#35bf28}+0.18\%$
test_a2c_speed[reduce-overhead-None] 1.1364ms 0.9448ms 1.0584 KOps/s 1.0529 KOps/s $\color{#35bf28}+0.52\%$
test_ppo_speed[False-None] 3.9864ms 3.8584ms 259.1743 Ops/s 265.3296 Ops/s $\color{#d91a1a}-2.32\%$
test_ppo_speed[False-backward] 7.3631ms 6.9342ms 144.2133 Ops/s 139.0408 Ops/s $\color{#35bf28}+3.72\%$
test_ppo_speed[True-None] 1.5562ms 1.3784ms 725.4967 Ops/s 710.9953 Ops/s $\color{#35bf28}+2.04\%$
test_ppo_speed[True-backward] 3.1557ms 2.9680ms 336.9322 Ops/s 336.9514 Ops/s $-0.01\%$
test_ppo_speed[reduce-overhead-None] 1.2028ms 0.9920ms 1.0080 KOps/s 965.5703 Ops/s $\color{#35bf28}+4.40\%$
test_reinforce_speed[False-None] 2.9343ms 2.2381ms 446.8033 Ops/s 442.1826 Ops/s $\color{#35bf28}+1.04\%$
test_reinforce_speed[False-backward] 3.7261ms 3.2792ms 304.9497 Ops/s 303.4403 Ops/s $\color{#35bf28}+0.50\%$
test_reinforce_speed[True-None] 1.3266ms 1.2080ms 827.7807 Ops/s 816.8569 Ops/s $\color{#35bf28}+1.34\%$
test_reinforce_speed[True-backward] 2.9084ms 2.8243ms 354.0756 Ops/s 335.1399 Ops/s $\textbf{\color{#35bf28}+5.65\%}$
test_reinforce_speed[reduce-overhead-None] 0.4740s 9.9059ms 100.9499 Ops/s 97.4444 Ops/s $\color{#35bf28}+3.60\%$
test_iql_speed[False-None] 9.9309ms 9.2548ms 108.0525 Ops/s 105.5688 Ops/s $\color{#35bf28}+2.35\%$
test_iql_speed[False-backward] 13.6883ms 12.9981ms 76.9343 Ops/s 74.0355 Ops/s $\color{#35bf28}+3.92\%$
test_iql_speed[True-None] 2.5349ms 2.0917ms 478.0845 Ops/s 464.6625 Ops/s $\color{#35bf28}+2.89\%$
test_iql_speed[True-backward] 5.0146ms 4.7060ms 212.4937 Ops/s 211.3857 Ops/s $\color{#35bf28}+0.52\%$
test_iql_speed[reduce-overhead-None] 16.6527ms 9.8936ms 101.0757 Ops/s 98.3833 Ops/s $\color{#35bf28}+2.74\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9172ms 5.7772ms 173.0956 Ops/s 173.3594 Ops/s $\color{#d91a1a}-0.15\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.0133ms 0.3465ms 2.8863 KOps/s 2.6954 KOps/s $\textbf{\color{#35bf28}+7.09\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6127ms 0.3292ms 3.0373 KOps/s 2.8339 KOps/s $\textbf{\color{#35bf28}+7.18\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.8298ms 5.5816ms 179.1607 Ops/s 178.9569 Ops/s $\color{#35bf28}+0.11\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9400ms 0.3674ms 2.7216 KOps/s 3.5743 KOps/s $\textbf{\color{#d91a1a}-23.86\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6001ms 0.3504ms 2.8542 KOps/s 3.8088 KOps/s $\textbf{\color{#d91a1a}-25.06\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6512ms 1.4140ms 707.2267 Ops/s 784.1558 Ops/s $\textbf{\color{#d91a1a}-9.81\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.6410ms 1.3421ms 745.0745 Ops/s 779.8060 Ops/s $\color{#d91a1a}-4.45\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 11.0051ms 5.8518ms 170.8880 Ops/s 173.8937 Ops/s $\color{#d91a1a}-1.73\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.2259ms 0.4527ms 2.2091 KOps/s 2.2096 KOps/s $\color{#d91a1a}-0.02\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7291ms 0.5005ms 1.9981 KOps/s 2.2969 KOps/s $\textbf{\color{#d91a1a}-13.01\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.6673ms 5.5871ms 178.9840 Ops/s 178.4186 Ops/s $\color{#35bf28}+0.32\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.0022ms 0.2892ms 3.4577 KOps/s 2.8710 KOps/s $\textbf{\color{#35bf28}+20.44\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4775ms 0.2713ms 3.6865 KOps/s 3.0813 KOps/s $\textbf{\color{#35bf28}+19.64\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.8702ms 5.5779ms 179.2800 Ops/s 178.7659 Ops/s $\color{#35bf28}+0.29\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.6853ms 0.3289ms 3.0409 KOps/s 3.5509 KOps/s $\textbf{\color{#d91a1a}-14.36\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6107ms 0.3063ms 3.2652 KOps/s 3.7950 KOps/s $\textbf{\color{#d91a1a}-13.96\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.1792ms 5.7920ms 172.6517 Ops/s 174.6070 Ops/s $\color{#d91a1a}-1.12\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1605ms 0.4876ms 2.0510 KOps/s 2.2705 KOps/s $\textbf{\color{#d91a1a}-9.67\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6790ms 0.4722ms 2.1179 KOps/s 2.3825 KOps/s $\textbf{\color{#d91a1a}-11.11\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.6241s 17.3545ms 57.6218 Ops/s 49.0852 Ops/s $\textbf{\color{#35bf28}+17.39\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 4.2522ms 1.8966ms 527.2490 Ops/s 474.5989 Ops/s $\textbf{\color{#35bf28}+11.09\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 2.1190ms 1.1546ms 866.0708 Ops/s 875.1252 Ops/s $\color{#d91a1a}-1.03\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.8431ms 5.1118ms 195.6261 Ops/s 197.7976 Ops/s $\color{#d91a1a}-1.10\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 4.1108ms 1.8433ms 542.4991 Ops/s 548.9140 Ops/s $\color{#d91a1a}-1.17\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 1.1044ms 0.9364ms 1.0680 KOps/s 1.0681 KOps/s $\color{#d91a1a}-0.02\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5577s 16.3103ms 61.3110 Ops/s 190.4976 Ops/s $\textbf{\color{#d91a1a}-67.82\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 4.0959ms 1.9590ms 510.4704 Ops/s 474.9331 Ops/s $\textbf{\color{#35bf28}+7.48\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.0526ms 1.0835ms 922.9102 Ops/s 823.1704 Ops/s $\textbf{\color{#35bf28}+12.12\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 37.7392ms 35.3819ms 28.2630 Ops/s 27.8239 Ops/s $\color{#35bf28}+1.58\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 20.0150ms 18.3534ms 54.4858 Ops/s 55.2033 Ops/s $\color{#d91a1a}-1.30\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 40.2971ms 36.7142ms 27.2374 Ops/s 26.7784 Ops/s $\color{#35bf28}+1.71\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 20.0655ms 18.6704ms 53.5606 Ops/s 53.6972 Ops/s $\color{#d91a1a}-0.25\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 40.6669ms 38.9207ms 25.6932 Ops/s 25.4755 Ops/s $\color{#35bf28}+0.85\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.3858ms 20.0065ms 49.9838 Ops/s 49.9598 Ops/s $\color{#35bf28}+0.05\%$

[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Jan 29, 2026
This commit adds a file-based auto-labeling system that complements the
existing title-based labeling (auto-tag.yml):

New files:
- .github/labeler.yml: Path-to-label mappings for all TorchRL components
- .github/workflows/auto-labeler.yml: Workflow using actions/labeler@v5

Updated files:
- .github/labels.yml: Added missing labels for Environments (gym, sklearn,
  dm_control, habitat), TorchRL modules (Objectives, Modules, Transforms,
  Record, Trainers, WeightUpdate, Services, ReplayBuffers), and other
  categories (sota-implementations/, tutorials/)
- .github/workflows/auto-tag.yml: Fixed to only ADD labels (never remove),
  added support for singular/plural prefix variations (e.g., [Environment]
  and [Environments] both work)

Together these provide complete PR categorization:
- Title-based labels capture the INTENT (bug fix, feature, etc.)
- File-based labels capture the SCOPE (which components changed)

ghstack-source-id: d1f282a
Pull-Request: #3402
@vmoens vmoens merged commit 5fe28cd into gh/vmoens/205/base Jan 29, 2026
61 of 81 checks passed
@vmoens vmoens deleted the gh/vmoens/205/head branch January 29, 2026 09:58
vmoens added a commit that referenced this pull request Jan 29, 2026
This commit adds a file-based auto-labeling system that complements the
existing title-based labeling (auto-tag.yml):

New files:
- .github/labeler.yml: Path-to-label mappings for all TorchRL components
- .github/workflows/auto-labeler.yml: Workflow using actions/labeler@v5

Updated files:
- .github/labels.yml: Added missing labels for Environments (gym, sklearn,
  dm_control, habitat), TorchRL modules (Objectives, Modules, Transforms,
  Record, Trainers, WeightUpdate, Services, ReplayBuffers), and other
  categories (sota-implementations/, tutorials/)
- .github/workflows/auto-tag.yml: Fixed to only ADD labels (never remove),
  added support for singular/plural prefix variations (e.g., [Environment]
  and [Environments] both work)

Together these provide complete PR categorization:
- Title-based labels capture the INTENT (bug fix, feature, etc.)
- File-based labels capture the SCOPE (which components changed)

ghstack-source-id: d1f282a
Pull-Request: #3402
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CI Has to do with CI setup (e.g. wheels & builds, tests...) CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant